Kimi K2.5 connects visual understanding with agents
Moonshot AI introduces native multimodality and coordinated parallel work.
Vibe 2.0 gives a coding agent working habits
Mistral adds configurable subagents, skills, and clarification to its terminal workflow.
Prism puts the manuscript and the model together
OpenAI introduces a LaTeX workspace for scientific writing and collaboration.
Claude’s constitution makes its intended values visible
Anthropic publishes a revised account of how it wants the model to behave.
AI education becomes an infrastructure question
OpenAI’s Edu for Countries initiative targets national education systems.
A year after DeepSeek, the open ecosystem is wider
Hugging Face looks beyond one breakout model to a changing community.
ChatGPT Go reaches a wider audience
OpenAI expands the subscription positioned between free access and higher tiers.
AI at work: measure tasks before predicting jobs
Anthropic’s Economic Index adds a more detailed view of how people use Claude.
The people shaping Claude’s next interfaces
Mike Krieger joins Anthropic Labs as the company separates experimentation from scaling.
Veo turns reference images into video ingredients
The update adds more control over visual continuity and vertical output.
Claude moves into healthcare workflows
Anthropic expands its healthcare and life-sciences tooling.
Tolan builds a voice companion around memory
An OpenAI case study explores speed, context, and continuity in conversation.
Falcon-H1-Arabic puts language first
TII’s model family focuses on Arabic rather than treating it as another checkbox.
Beyond the parameter count
A new architecture paper is a reminder that bigger is only one way a model can change.
An AI startup can begin before the idea
OpenAI’s next Grove cohort focuses on founders still finding their direction.
How to read an AI benchmark
Start the year with a better question: what, exactly, did the test measure?